Use of EM Methods for Groundwater Management
Rosemary Knight, March 2023

Q&A Session:

Q: Are your sediment type models specific to California or are they global? If specific, what's needed to adapt them to African soils?

A: We are trying to figure out whether they even work throughout California in general. Sands and gravels are more resistive. Clays are more conductive. But given the level of accuracy needed in the sediment type model, site-specific calibration will be required.

We are looking to use a well and calibration in the southern Central Valley and apply it in the northern Central Valley. There’s no way of knowing if this exercise can be successfully repeated in the African soils.

When I see geophysical data being acquired with no well data, no way to ground truth, no way to actually link what we measure electrical resistivity to sediment type, I get a bit concerned because you could have, for example, sands and gravels with a lot of pyrite magnetite (or Pyrrhotite) present. And that would make them nearly as conductive as clay.

Having not acquired any geophysical data in Africa, how to go about doing calibration is hard to answer. I'm cautiously exploring how transferable they are throughout California. And then Africa is a bit of a leap.

Additional comment: Soils can vary in acidity as well as iron content. So, you have the magnetic current ability to deal with it. Magnetite and pyrite are magnetic minerals that are in a lot of soils in the United States.

References:

Q: Have you been using AI for processing these huge data sets? As you know, AI/LM advantage is efficient handling large image sets.

A: No, we haven't been using AI processing. The algorithms are very robust. There are certainly opportunities for exploring AI applications. Now we have this streamlined process but there's so much insertion of decisions at various points. But again, the short answer to the use of AI is no, we haven’t.

Q: Does this mapping help just with assessing underground water? Does it say anything about soil quality as related to agriculture production?

A: Okay, so what we're getting is the electrical resistivity value of the top layer. So, I did some work, 20 years ago now, with David Robinson, now in the U. K., and we used one of the EM Induction tools to map out the electrical conductivity of the surface layer. And he was looking at this correlation between vegetation and electrical conductivity. And there was strong correlation between the zones where you had better vegetation and more electrical conductivity. And it was attributed to two reasons: more water and more clay. And the idea was that clays tend to retain the nutrients and water is needed. The problem with that argument in California has to do with the high electrical conductivity in the near surface layer that is very likely due to high salinity. So, there are definite applications for mapping out the conductivity of the soil layer and whether high electrical conductivity means “great,” or high electrical conductivity means “bad.” It's something we haven't explored yet.

Reference:

 Q: How quickly are you able to collect data along the tDEM for smaller scale?  What's the measurement spacing along the lines?

A: We just went down between the Almond Groves. So you can do whatever you want. It depends on what question you're trying to answer. We typically do either 10 feet or 20 feet between the rows and the size of the almond groves. We covered three almond groves like that in a day.

Reference:

Q: Do we need to sweep all the area to improve the accuracy of the Airborne EM method?

A: I really think it's important to figure out what is the question you want to answer before you start collecting data and this reconnaissance approach that Department of Water Resources took on and that I suggested is just to get a starting point. Can we acquire Airborne EM data everywhere? Can we, as a big picture, see where there is a sediment package that is coming off the mountains, where there are clay packages that extend depth. Now, depending on what I’m really interested in in an area, I will want to go back and start infilling between those lines. If I'm trying to locate a recharge pond with say a four-kilometer spacing; but this reconnaissance scale spacing isn't going to be adequate for detailed groundwater management. So, what I suggest is you come up with your model and you say, where are my predictions that are important for making this decision really impacted by lack of data? So, you started to a sensitivity analysis and say, where is the uncertainty in my model creating so much uncertainty in my predictions that I really don't feel comfortable moving forward. And then you go on to acquire data in those “gap” areas.

Q: In your experience, have you observed the AEM method identifying changes in subsurface conditions from large development projects?

A: No, we haven't looked at that yet. Now, part of the issue with Airborne EM, is as you cannot fly over a populated area and that's because there is a non-zero risk that the transmitter loop’s going to fall off. So, you don't want to be in a populated area, and you cannot fly over developed areas. Something I’m passionate about right now and starting to work with people at the state level on this is, I'm calling it “nature first land use planning” or “nature 1st zoning”. That is, you don't allow a development project in an area that is best suited for groundwater recharge. You think about how to use your physical data to understand the subsurface. You use your geophysical data to understand how nature is operating here. And we think about the natural infrastructure that we have, not just the engineering infrastructure, but we think about how we use the natural infrastructure to tie into our engineered infrastructure to really support sustainable groundwater management. So, I'm very interested in predicting changes in subsurface conditions from large development projects before the development project goes in.

Q: Can anyone sign up for the course on how to use this assessment tool? if so, what is the likelihood that data can be collected elsewhere, and use this tool to do the same mappings? Will the tool work on other data sets that are acquired in other areas in the U.S. and in other countries?

A: Unfortunately, I think all the data formatting is specific to the California case. If there is sufficient interest, like in Africa, anywhere else in the world, I would take this to the school and see if we can get additional funding to partner with someone to develop an interface that could then accommodate different data formats that might be useful.

Additional comments: in a lot of these African countries and places like Navajo Nation, there have been no surveys. So even assistance with this is how you collect the data, this is the format that would actually work in the GEM software tool. Assuming your assessment tool is available to team up with other universities across the country and world.

Q: How often is the mapping needed to be repeated for same surface?

A: What we're mapping out is primarily the distribution of sediment type. So, if you just want that, that's not going to change. You just need to do it once. Along the coast, however, we've done repeat surveys to look at saltwater intrusion. And this method is absolutely spectacular. I was running out of time, so l cut the saltwater intrusion example. But you just see this gorgeous saltwater wedge with fresh water over the top. And so, the agency has now paid for repeat surveys there three years apart. Didn't see a huge difference, actually. And we are just now submitting a proposal asking for funding to try to repeat airborne EM surveys in California to see if we can detect the change in stored water. So that's something we haven't done yet.

References:

Q: Can we plug-in electromagnetic equipment in UAVs to lower the cost of the skyTEM? Like removing the helicopter and use UAVs instead.

A: That would be a great idea. There are people exploring drones, but the transmitters are too heavy for a drone. But what we did with the drone, and it was a collaboration with a group out of Japan, they put the transmitter around the Almond Grove and then flew the receiver. It was a great experiment, but the signal to noise ratio was very low. At this point in time, no one's explored the whole UAV idea.

Reference:

Q: Could this method capture the materials ("sediment types") that we can find in the soil (and in the undersoil)? Also, can AEM be efficient to detect groundwater resources in arid regions and desert.

A: I'm not sure what kind of materials. It can detect anything where there's an electrical resistivity contrast. So, any sediment type.

The EM method can be efficient to detect groundwater resources in our region. Yes, absolutely. The main message of this talk is a very powerful method.

References:

Q: Do the long periods of drought make a difference in the structure of the soil and that will affect how we recharge the underground reservoirs.?

A: The effect of the drought and this subsidence that occurs can cause irreversible compaction. When you pump, the sands don't change much but the clays collapse when it gets below the hydraulic head value. That has definitely happened in parts of the San Joaquin Valley and that means there is a loss of storage. So even if you recharge, you're never again going to pop open those clays and fill up that component of the storage.

I'm forgetting what percent of the subsidence. It's a concern but we can't really detect it with the airborne EM but do detect it with the inSAR by mapping of elevation change and then we model it.

References:

Q: Could this method help us to save the natural water table in the desert?

A: This AEM method would help you understand the current situation and could then help design plans so that the water table doesn't continue to be lowered.

The big problem with groundwater is out of sight, out of mind. If you're pumping your water out of a lake and you see the lake level going down, down, down, down, you realize you've got a problem.

If you pump groundwater and you don't see anything happening, you don't realize you've got a problem until your well goes dry. So that is where I think the value is. This method helps us understand. Also, if I put a well here and pump, pump, pump, is it going to impact my neighbor’s well over there? So AEM method helps us understand the connections between what we do at the surface, what we do in our well and the larger scale impact we're having on the groundwater system.

Reference:

Q: This was a great presentation, and very informative! Nature First!

A: OK. So, I see I've got buy in for my Nature First approach. Yeah, that nature first has to be the approach to everything. Let's not mess it up. Let's understand the natural system and think of that as we design our engineered systems. Because nature is even better than our best engineers when it comes to finding ways to protect our water resources.

References: